| Author |
|
smarty
Super User
Joined: May 21 2006
Location: United States
Online Status: Offline
Posts: 728
|
| Posted: March 02 2015 at 19:12 | IP Logged
|
|
|
I have been seeing an intermittent issue where PH crashes (just vanishes, no on screen errors, no warning, just disappears).
By looking at the logs after a re-start, it appears that this occurs when I have a trigger that fires when my Itach controller goes offline (CONTROLLER STAT1). My Itach controller is a Wi-Fi to IR unit. Recall that there were recent changes to PH to make the Itach controller restart more robust. Maybe my re-start macro is trampling something? My Wi-Fi is decently stable, but I see this issue about once every 2-3 days.
See the two separate log occurrences below:
Note that in this one, I added some user log entries into the offending macro to see what line of the macro failed (the "traps"):
Here are the triggers that could call the macro:
Here is the macro that is called. Note where I added the traps and compare to the last log file:
It appears that the macro stopped executing at line 20. Not sure why. Line 20 disables the triggers so they don't keep continually firing during my attempt at reconnecting the controller.
Ideas? Is there a better way to do this? I need to have my Itach stay connected, or, a reliable way of having it reconnect without PH crashing.
__________________
Elk - Insteon - BlueIris - DMC1 - PowerHome - XLobby - HA_Bridge w/Dots - Brultech
|
| Back to Top |
|
| |
dhoward
Admin Group
Joined: June 29 2001
Location: United States
Online Status: Offline
Posts: 4447
|
| Posted: March 03 2015 at 20:23 | IP Logged
|
|
|
Steve,
If you're on the latest version of PowerHome (2.1.5c)
try just disabling the trigger. I seem to have the same
problem with my Global cache as well. The new code
should automatically handle reconnects without you
having to do it in a trigger.
Give it a shot and let me know if doesnt auto-reconnect.
Dave.
|
| Back to Top |
|
| |
smarty
Super User
Joined: May 21 2006
Location: United States
Online Status: Offline
Posts: 728
|
| Posted: March 05 2015 at 06:57 | IP Logged
|
|
|
Yes running latest 2.1.5c.
I disabled both triggers....after a full boot & restart. PH ran again for about 1.5 days, then crashed as above. Without the trigger being enabled, I was wondering if I could track it back to my Itach.
After the PH crash, I restarted PH (without boot). It started normally, except for this:
The log files show me nothing other than the new ECM-1240 trigger I am working on (I am pretty sure this is NOT the cause):
Hope this helps track this bug.....
__________________
Elk - Insteon - BlueIris - DMC1 - PowerHome - XLobby - HA_Bridge w/Dots - Brultech
|
| Back to Top |
|
| |
dhoward
Admin Group
Joined: June 29 2001
Location: United States
Online Status: Offline
Posts: 4447
|
| Posted: March 05 2015 at 21:11 | IP Logged
|
|
|
Steve,
Im pretty sure its not something caused by you. Im
going to have to revisit the Global cache
failure/reconnect code and make some changes.
Do you remember when was the last version that worked
and the first version where it was broken? It'll help
me isolate what changed so I can get it patched.
Thanks,
Dave.
|
| Back to Top |
|
| |
smarty
Super User
Joined: May 21 2006
Location: United States
Online Status: Offline
Posts: 728
|
| Posted: March 06 2015 at 08:23 | IP Logged
|
|
|
Dave,
My Itach Wi-Fi2IR set up has had a bit of a bumpy past.
In searching back through the PH forum, on 8/8/14 you were assisting me with troubleshooting virtual serial port issues (Itach GLOBAL CACHE issues, NOT Itach Wi-Fi2IR. You mentioned that you changed over to Catalyst Controls as an upgrade to help the Global Cache disconnect/reconnect issue - not sure that any of this applies to the Wi-Fi2IR (that module doesn't use a PH-VSP like the Global caches does).
see here: (Page 8 and 9) http://www.myx10.com/forum/forum_posts.asp?TID=3384&KW=itach +smarty&PN=0&TPN=8
As part of the saga, I changed out my serial card to a Win 7 certified card (issues with my Global Cache - not Itach Wi-Fi2IR).
So...To answer your question....WiFi2IR issues to be seen somewhere in the 2.1.5a-c range (although not sure it ever was fully stable because I there were other issues ongoing at the same time - and that was about the time I got the unit).
Not much help 
Sorry
__________________
Elk - Insteon - BlueIris - DMC1 - PowerHome - XLobby - HA_Bridge w/Dots - Brultech
|
| Back to Top |
|
| |
smarty
Super User
Joined: May 21 2006
Location: United States
Online Status: Offline
Posts: 728
|
| Posted: March 24 2015 at 19:37 | IP Logged
|
|
|
I think I have new/additional information to provide on this subject...where PH just crashes....
By using a trigger that fires a macro (the macro has only one line - it just records the TEMP variables that are past from the trigger... and I have them recorded to the event log) when my iTach WiFi2IR controller DISCONNECTS, I can tell that the PH crash occurs as PH tries to reconnect to the iTach controller. See image below:
]
The above information is not new. What I believe is new follows:
Note - I have two physical Global Cache brand devices. One is my GC-100 (it has two serial ports that require two instances of the PH_VSP.phvsp plugin to use them). The second device is a WiFi2Ir unit. It is the controller for this WiFi2Ir unit that periodically disconnects (causes the trigger to fire).
After PH crashes, Windows and my other running programs still appear OK. However, if I restart PH (without doing a windows boot), I find that PH has an error when it tries to load the PH_VSP.phvsp plugins. See the image below:
If I then try to reinit or restart PH again (without the windows boot), I ALWAYS have this plugin loading error during PH startup. The ONLY way to have PH start and load everything with no errors requires a windows boot.
So, it is my thinking that maybe something is happening with the PH_VSP.phvsp either before or after the PH crash... Just another clue....
__________________
Elk - Insteon - BlueIris - DMC1 - PowerHome - XLobby - HA_Bridge w/Dots - Brultech
|
| Back to Top |
|
| |
smarty
Super User
Joined: May 21 2006
Location: United States
Online Status: Offline
Posts: 728
|
| Posted: March 24 2015 at 19:38 | IP Logged
|
|
|
For now, to keep my system stable, I just leave the controller for the WiFi2IR device "unchecked" until it is needed.
__________________
Elk - Insteon - BlueIris - DMC1 - PowerHome - XLobby - HA_Bridge w/Dots - Brultech
|
| Back to Top |
|
| |
smarty
Super User
Joined: May 21 2006
Location: United States
Online Status: Offline
Posts: 728
|
| Posted: April 02 2015 at 07:50 | IP Logged
|
|
|
Update - With controller unchecked - the PH has been rock stable since then.
__________________
Elk - Insteon - BlueIris - DMC1 - PowerHome - XLobby - HA_Bridge w/Dots - Brultech
|
| Back to Top |
|
| |
dhoward
Admin Group
Joined: June 29 2001
Location: United States
Online Status: Offline
Posts: 4447
|
| Posted: April 02 2015 at 23:59 | IP Logged
|
|
|
Steve,
I'll be reworking the reconnect code for the Global
Cache controller (the WiFi2IR). That is what is causing
PowerHome to crash and consequently cause the VSP
plugins to hang (when PH crashes, it does not properly
unload the VSP system hooks. When you relaunch
PowerHome, the VSP hooks are still residual so you mush
restart the PC to clear this condition).
Dave.
|
| Back to Top |
|
| |
smarty
Super User
Joined: May 21 2006
Location: United States
Online Status: Offline
Posts: 728
|
| Posted: August 07 2015 at 07:56 | IP Logged
|
|
|
Dave,
After doing some upgrades to my wireless network, I have decided to move away from the iTach WiFi2IR device because it only works on the wireless "b" network.
Most Wireless Access Points (WAP's) only have a single radio; because of this, the speed of the WAP down shifts to your slowest wifi device (the "b" speed of the WiFi2IR device).
Therefore, I will be moving to wired CC-100-06 and this not have:
1)the wifi network speed reduction, and
2)the disconnect/reconnect PH hang up problem
I hope you have made it back home and all is well. We miss your interaction on the forum.
Steve
__________________
Elk - Insteon - BlueIris - DMC1 - PowerHome - XLobby - HA_Bridge w/Dots - Brultech
|
| Back to Top |
|
| |
dhoward
Admin Group
Joined: June 29 2001
Location: United States
Online Status: Offline
Posts: 4447
|
| Posted: August 12 2015 at 19:35 | IP Logged
|
|
|
Steve,
Yes! I made it back safely  . While I was gone, my
trusty Global Cache GC-100-12 died on me. Everything
appears fine (hard resets work properly, etc), just
cant communicate to it via TCP/IP. Something must have
fried in the communication circuits.
Anyways, I had to resort to using my iTach WiFi2IR
device as well and saw exactly what you were seeing. I
also ordered a used GC-100-6 from eBay but it was a
later revision than my original GC. It also had
problems. The good news is that with all this first
hand access to the problems, I believe Ive finally got
the GC code fixed where its working properly with the
WiFI2IR and the later revision GC-100-XX devices.
Consequently, I'll be releasing a fix6 here shortly
which will have all my latest code. Ive also made a
couple more small tweaks to the VSP plugin concerning
auto-reconnect so hopefully that will be a little more
stable as well.
Stay tuned as I should have the fixes out within a
week.
Dave.
|
| Back to Top |
|
| |
smarty
Super User
Joined: May 21 2006
Location: United States
Online Status: Offline
Posts: 728
|
| Posted: November 25 2015 at 13:13 | IP Logged
|
|
|
Hmm...Not sure how I could get 10 out 8 controllers connected....
Running 2.15d...I see this fairly regularly after a day or two without a reboot.
Edited by smarty - November 25 2015 at 13:15
__________________
Elk - Insteon - BlueIris - DMC1 - PowerHome - XLobby - HA_Bridge w/Dots - Brultech
|
| Back to Top |
|
| |
dhoward
Admin Group
Joined: June 29 2001
Location: United States
Online Status: Offline
Posts: 4447
|
| Posted: November 29 2015 at 14:31 | IP Logged
|
|
|
Steve,
Interesting...I havent seen this yet. Do you regularly
disconnect/connect individual controllers? If so, what
method do you typically use?
Ive added this to the bug list but let me know on the
above as it will help in tracking it down.
Thanks,
Dave.
|
| Back to Top |
|
| |
smarty
Super User
Joined: May 21 2006
Location: United States
Online Status: Offline
Posts: 728
|
| Posted: November 30 2015 at 10:19 | IP Logged
|
|
|
Dave,
To answer your questions the best that I can....I do NOT regularly disconnect/connect my PH controllers.
Other than a GlobalCache that is connect via a Wi-Fi bridge, all of my system is hardwired. I suspect that periodically the Wi-Fi bridge losses connectivity to the WAP...which possibly leads to the CG becoming disconnected. To potentially help this, I have went through all the power and driver settings so that these devices NEVER power down or "go to sleep". However, I still see this unexpected message after a few days.
Edited by smarty - November 30 2015 at 10:20
__________________
Elk - Insteon - BlueIris - DMC1 - PowerHome - XLobby - HA_Bridge w/Dots - Brultech
|
| Back to Top |
|
| |
BeachBum
Super User
Joined: April 11 2007
Location: United States
Online Status: Offline
Posts: 1880
|
| Posted: November 30 2015 at 15:33 | IP Logged
|
|
|
I hate to throw gas on the fire but I have a similar problem with my USB PLM. If I have a PLM failure I get a system message trying to recover then I get a box with a message that I can't get rid of except by Task Manager. Meanwhile PH terminates and unless I end task in the box I'm hung. May be totally a separate issue but it is a rock solid failure.
__________________
Pete - X10 Oldie
|
| Back to Top |
|
| |
dhoward
Admin Group
Joined: June 29 2001
Location: United States
Online Status: Offline
Posts: 4447
|
| Posted: November 30 2015 at 15:57 | IP Logged
|
|
|
Pete,
Can you get me screenshots of the various messages and
windows? Its a totally separate problem (I suspect
Steve's problem is the menu controller count is getting
off when the Global Cache auto-reconnects) but one that
still needs to be fixed.
Thanks,
Dave.
|
| Back to Top |
|
| |
BeachBum
Super User
Joined: April 11 2007
Location: United States
Online Status: Offline
Posts: 1880
|
| Posted: December 04 2015 at 06:12 | IP Logged
|
|
|
Dave,
Sorry about the delay.. Will do over the weekend.
__________________
Pete - X10 Oldie
|
| Back to Top |
|
| |
smarty
Super User
Joined: May 21 2006
Location: United States
Online Status: Offline
Posts: 728
|
| Posted: February 20 2016 at 11:09 | IP Logged
|
|
|
An update on "Controllers Connected" count....
Dave - I have now (as of a number of months back) hardwired my Global Cache (and all other PH controllers). I do still see my controller count get out of whack. Presently, it is at... "Controllers Connected (12 of 8)". While this looks odd, it doesn't cause any real issues :)
As an additional observation, even with my Insteon USB stick as my Insteon controller; with high activity (like when we are have a house load of guests...(many lights on and various dimmed scene settings)... I see PH trying to periodically auto-reconnect to the Insteon controller.
Therefore, I suspect two things:
1) The controller count issue is driven my Insteon controller trying to reconnect, not my Global Cache.
2) Something is still amiss with the Insteon controller (only under heavy stress usage) where PH recovers, but only by doing the re-connect attempts. Recall I am running an updated, FRESH install of Win 7 x32 and fresh install of PH 2.1.5.d
__________________
Elk - Insteon - BlueIris - DMC1 - PowerHome - XLobby - HA_Bridge w/Dots - Brultech
|
| Back to Top |
|
| |
dhoward
Admin Group
Joined: June 29 2001
Location: United States
Online Status: Offline
Posts: 4447
|
| Posted: February 22 2016 at 20:52 | IP Logged
|
|
|
Steve,
Thanks for the feedback. You're probably right that its
the Insteon controller that is the main contributor.
Based upon your feedback, Pete's, and my own internal
testing, Ive been redoing the internals of the Insteon
controller yet again. Hopefully I'll get it nailed this
time.
Dave.
|
| Back to Top |
|
| |
Active Topics 
Memberlist 
Search 
Help
Register 
Login
PowerHome Messageboard : PowerHome Bug Reports
Topic: PH Crashing - Controller Restart
